Microbiome and virome on indoor surfaces of an Antarctic research ship

BACKGROUND Few studies have focused on microbial diversity in indoor environments of ships, as well as the role of the microbiome and its ecological interconnections. In this study, we investigated the microbiome and virome present on the internal surfaces of a polar ship in different stages (beginning, during, and at the end) of the Brazilian Antarctic expedition in order to evaluate abundance of microorganisms in different periods. OBJECTIVES AND METHODS We used shotgun metagenomic analysis on pooled samples from sampling surfaces in the ship’s interior to track the microbial diversity. FINDINGS Considering the total fraction of the microbiome, the relative abundance of bacteria, eukaryotes, viruses, and archaea was 83.7%, 16.2%, 0.04%, and 0.002%, respectively. Proteobacteria was the most abundant bacterial phyla, followed by Firmicutes, Actinobacteria, and Bacteroidetes. Concerning the virome, the greatest richness of viral species was identified during the middle of the trip, including ten viral families after de novo assembly: Autographiviridae, Chrysoviridae, Genomoviridae, Herelleviridae, Myoviridae, Partitiviridae, Podoviridae, Potyviridae, Siphoviridae, and Virgaviridae. MAIN CONCLUSIONS This study contributed to the knowledge of microbial diversity in naval transportation facilities, and variations in the abundance of microorganisms probably occurred due to factors such as the number of passengers and activities on the ship.

Ships are semi-closed and densely populated environments of close living and sleeping quarters, shared water, meals, and ventilation and sewage systems. (1)Because of these specificities, ships are considered potential sources of disease outbreaks, and promoters of transmission of pathogens already present or introduced on board.Therefore, the spread of vectors and microorganisms from crew and passengers to land-based populations and vice versa should be considered. (2,3,4,5)This last assertion is particularly relevant considering expeditions to the Antarctic continent, due to the risk of introducing and spreading exotic or autochthonous species in a relatively unexplored and untouched continent.
Microbial infections are probably the most common acquired diseases indoors, mainly in poorly ventilated environments. (6,7,8)Pathogens such as norovirus, influenza virus, Legionella spp., Salmonella spp., E. coli, Vibrio spp., Mycoplasma pneumoniae, as well as vaccine-preventable diseases such as measles, rubella, and varicella are well documented in cruise ships and military vessels. (2,3,9,10,11,12)More recently, the transmission of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in cruise and military ships, among other transportation facilities, has been well documented (4,5,13) with implications not only for the health of the crew but also contributing to the spread of SARS-CoV-2 among different coastal locations and continents. (13)n the context of a polar expedition, the researchers also collect samples from the soil, sea, lakes, ice, and animal excrement.They can transport microorganisms in the polar ecosystem to the ship through contact with these matrices.The primary issue is due to the dispersal and adaptive capacities of some organisms, mainly pathogens, to the other continents and eventually introducing ecological and public health risks.
Microorganisms are deposited in the environment when excreta or other secretions (such as mucus, saliva, urine, and faeces) containing high concentrations or viral titres are released from an infected individual. (14,15,16)or instance, faeces can contain up to 10 12 viral particles per gram and vomit up to 10 7 per millilitre, so the potential cross-contamination from hands to surfaces is considerable. (1,14)Moreover, other significant sources of indoor microorganisms may be human oral and respiratory fluid emitted via coughing, sneezing, talking, and breathing or the direct shedding of skin-associated microbiota. (6,7,17)Toilet flushing can also aerosolise significant concentrations of viruses. (1)iruses can be transmitted through person-to-person contact or waterborne, foodborne, airborne, and vectorborne.In addition, the high stability of viruses or other microorganisms on surfaces or fomites emphasises the possible role of surfaces in the transmission route, mainly through contact. (3,14,16,18,19)Both enveloped and nonenveloped viruses are readily transferred between fomites and fingerpads, with an estimated transfer rate of ~22%. (16)Therefore, fomites are essential vehicles for the spread of pathogens and associated diseases. (14,20,21,22)n addition to pathogens that may be present on surfaces, recent studies have demonstrated that indoor surfaces of transportation facilities are microbial reservoirs from multiple sources.Human presence and nearby surroundings contribute to the characteristics of the microbiome. (6,7,8,23,24,25)Some factors can influence the microbial composition on indoor surfaces including material types, moisture rates, temperature, cleaning practices, human occupancy, and occupant activities. (6,23,24,25,26)owever, more information is needed about microbiome variations on indoor surfaces in transportation facilities under different physical-chemical conditions, types of materials, transport routes, number of passengers, distinct activities, and geographic location. (6,8)n this study, we analysed the microbiome and virome, including target viruses with RNA (ribonucleic acid) genome, from swab samples collected on indoor environmental surfaces of a Brazilian Navy Polar vessel (Almirante Maximiano -H41).For this, we used a metagenomic approach to track the microbial diversity during different stages of an expedition to Antarctica, considering other activities and the number of passengers on the ship.Swab samples collected on several surfaces during various expedition stages were pooled to assess the abundance of microorganisms present at each stage and whether autochthonous or allochthones environmental species could be carried between different continents.

MATERIALS AND METHODS
Sampling -Ninety-one swab samples from fomites and surfaces were obtained from different indoor environments (Fig. 1, Table I) according to a protocol by Ganime et al., (19) with minor modifications.Briefly, samples were obtained by swabbing at least 50% of a selected surface area with rayon swabs dipped and stored in 2.0 mL Dulbecco′s Phosphate Buffered Saline (PBS, 1.5X), pH 7.2.Swab samples were collected in the same places at three different periods during the Brazilian Antarctic expedition in 2019/2020 (Fig. 1).The first collection took place on October 7th, 2019, in Rio de Janeiro port, before the departure of the ship for the expedition (n = 34 samples); the second on February 3rd, 2020, while crossing the Drake Passage and anchoring in the port of Punta Arenas, Chile (n = 32 samples); and the third on April 5th, 2020, when the ship arrived in the port of Rio de Janeiro, Brazil, at the end of the expedition (n = 25 samples) (Fig. 1, Table I).These three periods were chosen to assess the microbial communities before the ship's departure, during the expedition, when a more significant number of people were on board, with frequent changes of part of the passengers, and at the end of the expedition to assess the potential microbial spread in the indoor environment.At the beginning and end of the trip, when the ship is moored, there is no crew on board, only a few navy officers responsible for the ship's maintenance.During the expedition, about 110 crew members were on board (https://www.marinha.mil.br/navio-polar-almirante-maximiano), including researchers and navy personnel.Before the voyage, the ship's surfaces are decontaminated to receive the crew.At the end of the expedition, the vessel underwent a surface decontamination process before docking in Rio de Janeiro, Brazil.Such actions were intensified after the decree made by the World Health Organization (WHO) in March 2020 of the coronavirus disease 19 (COVID-19) pandemic.
Surface samples were collected in places of greater crew circulation, such as cabins, restrooms, and handrails, where people touched more frequently (Fig. 1, Table I).
After swab collection, samples were processed immediately (stages 1 and 3) or kept under refrigeration at 4ºC during the trip (stage 2), until processed at the Laboratory of Comparative and Environmental Virology at Instituto Oswaldo Cruz (FIOCRUZ/RJ).In all periods, one swab sample was collected per location indicated in Table I, except for some examples collected in duplicate at the beginning of the expedition (October 7th, 2019) in specific fomites, such as faucets, cabinets, and freezer knobs.However, in subsequent collections, we considered it unnecessary to sample in duplicate.For metagenomics analysis, samples collected in each sampling period were pooled (10 µL from each sample kept in PBS solution) totalling three pools (named pool 1: collection carried out on October 7th, 2019 (n = 34); pool 2: collection carried out on February 3rd, 2020 (n = 32), and pool 3: collection carried out on April 5th, 2020 (n = 25).
Nucleic acid extraction, reverse transcription, preparation of genomic libraries, and sequencing -Samples were prepared according to the protocol described by Fernandez-Cassi et al. (27) Briefly, 150 μL of each pooled sample was treated with 160 U of Turbo DNAse (Ambion Cat no.AM1907, Ambion) for 1 h at 37ºC to remove free DNA (Deoxyribonucleic acid).DNAse was inactivated using the provided inactivation reagent, and the samples were centrifuged at 10,000 ×g for 1.5 min.The treated supernatants were collected, and 140 µL of the DNAse treated samples were extracted using the QIAamp ® Viral RNA Mini kit (QIAGEN, CA, USA) in a QIAcube ® automated system (QIAGEN) without the addition of RNA carrier.RNA templates were reverse transcribed precisely as Fernandez-Cassi et al. (27) to identify viruses with RNA genomes.The second cDNA strand construc-tion, and a polymerase chain reaction (PCR) amplification step were performed to obtain sufficient DNA for library preparation. (27,28)PCR products were purified, and concentrated to a volume of 50 µL using the Agencourt AMPure XP PCR purification kit (Beckman Coulter, CA, USA).Negative controls (DNase/RNase free water) were included in all stages of the procedures, and agarose gel electrophoresis was performed to verify DNA amplification, as described by Wang et al. (28) The purified amplicons were quantified using Qubit 2.0, and DNA libraries were generated using a Nextera XT DNA Preparation Kit (Illumina, San Diego, CA, USA).The size distribution of the libraries was evaluated using a 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA), and DNA High Sensitivity quantification was obtained using a Qubit 4.0 Fluorometer.Paired-end sequencing (2 x 150 bp) was performed using the Next-Seq platform (Illumina, San Diego, CA, USA) at SENAI CETIQT's Facility (SENAI Innovation Institute for Biosynthetics, Technology Centre and Textile Industry, Rio de Janeiro, RJ, Brazil) and PhiX was used as a control for Illumina sequencing runs.
Bioinformatics and data analysis -The reads in FASTQ format were generated by the Illumina Bas-eSpace pipeline (https://basespace.illumina.com).Lowquality sequences were filtered (Phred score < 20), and adapters were removed with trimmomatic v0.39. (29)The read quality analyses were performed using FastQC v0.11.9 (30) before and after trimming.The software Krak-en2 was used for a metagenomic and taxonomic assignment using the standard Kraken2_DB database. (31)The relative abundance estimates of microorganisms were refined using Bracken software. (32)Classified microorganisms with less than 10 reads were excluded from the relative abundance analysis.
Reads were de novo assembled using metaSPAdes (33) for paired-end reads.Contigs longer than 150 bp were queried for sequence similarity search using Blastx (parameters: e-value 1e-10 -max_target_seqs 25) (34) against the NCBI RefSeq database.Subsequently, BASTA (Basic Sequence Taxonomy Annotation) (parameters: -m 1 -l 75 -i 70), (35) was used to determine the taxonomy annotation of Blastx hits based on a last common ancestor algorithm.The species' nomenclature and classification were according to the NCBI (National Centre for Biotechnology Information) Taxonomy database standards (https://www.ncbi.nlm.nih.gov/guide/taxonomy/).Each viral contig (≥ 150 bp) was manually checked through Blastx, and the predicted viral hosts were inferred based on the closest relative sequence (lower evalue and higher score and nucleotide identity) found in the database by protein alignment along with information reported by the International Committee on Taxonomy of Viruses (ICTV -Report on Virus Classification and Taxon Nomenclature) (https://talk.ictvonline.org/ictv-reports/ictv_online_report/).Data availability -Raw reads are publicly available in the Sequence Read Archive (SRA) (NCBI -https:// www.ncbi.nlm.nih.gov/sra)individually with accession numbers (SRX15809764 to SRX15809766) under Bio-Project accession number (PRJNA850925).

RESULTS
Microbial diversity -A total of 19,502,486 pairedend reads were obtained from the sequencing of the three pools of samples.Table II shows the number of reads before and after trimming, low-quality sequences filtering, and the percentage of reads obtained for each biology domain obtained in each metagenomic library after taxonomic classification.
The analysis shows dominance of bacteria in the pooled samples corresponding to the beginning (stage 1) and the end of the expedition (stage 3).In contrast, pool 2, representing samples from the fully crewed ship, the Eukarya domain was dominant (Table II).
Some bacterial species related to the Antarctic continent were observed in the samples collected during the different stages of the expedition (Table III).Table III presents the results of the taxonomic classification obtained for other bacterial species related to the Antarctic environment and their relative abundance by pooled sample.The relative abundance of bacteria associated with the Antarctic was very low concerning the total quantity found for other species (Table III).Regarding the Eukarya domain and the Fungi Kingdom, two phyla were identified in all metagenomic libraries: Basidiomycota and Ascomycota.These two phyla corresponded to 0.05%, 22.8%, and 0.25% of the total reads obtained in stages 1, 2, and 3, respectively (Fig. 3A).Basidiomycota was identified in low abundance considering all reads assigned in stages 1, 2 and 3 (0.01%, 0.1% and 0.13%, respectively).Ascomycota represented 22.7% of all reads set in stage 2, with low abundance in stages 1 and 3 (0.03% and 0.11%, respectively).
Reads were reassembled to refine viral analysis using metaSPAdes (Table IV).Table IV shows the description, distribution by length and similarity search using Blastx of the assembled reads per sample.A higher number of contigs was observed in stage 2 compared to stages 1 and 3 (Table IV).In total, 38 viral contigs were assembled.
In stage 1, Propionibacterium phage PHL041M10 of the Pahexavirus genus and Actinomyces virus Av1 belonging to genera Dybvigvirus were identified (Table V).A greater richness of viral species was identified in samples from stage 2, collected four months after the beginning of the expedition (Table V).Single-strand RNA viruses (ssRNA+) that infect plants (represented by the Tobacco mosaic virus from the Virgaviridae family) were predominant in this pooled sample (Table V).Watermelon mosaic virus (Potyviridae family, Potyvirus genus, ssRNA+) was also identified, besides other plant (vegetal/fruit)-infecting viruses, such as Tomato brown rugose fruit virus and Pepper Mild Mottle Virus (PMMoV) (Table V).We also identified some species of the Pahexavirus genus, which are dsDNA bacteriophages with the predicted hosts (Propionibacterium sp., and Propionibacterium acnes).Data on Staphylococcus-infecting phages in stage 2 reflected the presence of Staphylococcus sp. in this pooled sample, corroborating microbiome findings.Another member of the Siphoviridae family (Psychrobacter phage Psymv2) was identified, containing sequences closely related to an open reading frame (ORF) coding a phage head-tail connector (Ref-Seq YP_009017594.1)(Table V).Flavobacterium infecting phages and viruses associated with phytopathogens of potato (Dickeya phage vB_DsoP_JA10 and Dickeya phage Ninurta) were observed, beyond enterobacteria infecting phages (Yersinia phage phiR1-37) and Acinetobacter virus Acj61 (Table V) infecting the predicted host Acinetobacter johnsonii found in aquatic sources, human skin, and animals.
At the end of the expedition, a predominance of fungiinfecting viruses (dsRNA) of the genus Alphachrysovirus (Chrysoviridae family) was observed, as well as a member of the Partitiviridae family and PMMoV (Table V).

DISCUSSION
Metagenomic data -In this study, we aimed to evaluate the microbial diversity present on indoor surfaces of a polar ship in different stages of an expedition from Brazil to the Antarctic continent, using a shotgun metagenomic approach.The main results demonstrated that the most abundant bacterial phyla corresponding to the total fraction of microbiome were Proteobacteria, followed by Firmicutes, Actinobacteria, and Bacteroidetes and according to other studies conducted to evaluate the microbiome in indoor surfaces of built environments or transportation facilities. (24,36,37)However, it was possible to observe variations in the relative abundance of microbial families and genera during different expedition stages.At the beginning and end of the expedition, with a limited number of passengers on the ship, it was possible to observe the prevalence of bacterial genera commonly present in the environment, such as Pseudomonas spp.(stage 1) and Massilia spp.(stage 3).Pseudomonas is a gram-negative, ubiquitous bacteria (widely found in diverse environments) of Pseudomonadaceae.Genera Massilia sp. are psychrophilic or mesophilic and are generally considered environmental organisms rather than animal-associated and have already been isolated from different settings, such as freshwater, glaciers, rocks, and air samples. (38)different profile of bacterial genera was observed in stage 2 (during the expedition).Cutibacterium was the most relatively abundant microbial genus in stage 2, and similar findings were observed by Danko et al., (24) which identified a higher abundance of Cutibacterium acnes (known human skin commensal) in indoor surfaces of urban transportation facilities.Staphylococcus was the second bacterial genus most prevalent in this stage and comprises gram-positive bacteria that can be found both on the human skin and on the nasal mucous.The abundance of bacterial genus commonly present in indoor surfaces of built environments or transportation facilities with human presence has been documented. (23,24,37)he top taxa associated with indoor environments were recognisable as microbes associated with humans (e.g., Corynebacterium, Streptococcus, Enterobacteriaceae, Staphylococcus, Propionibacterium, Lactococcus) (23) and are also comparable with our results, corroborating the findings of our study for pooled samples of the middle of the expedition, when passengers were confined.This distinct pattern may be related to the more significant number and circulation of people and activities on the ship during the middle of the expedition since the nature of human contact and human behaviour highly influences indoor surfaces. (6,8,23)t is worth noting that samples from stage 2 (collected during the expedition) were stored at 4ºC before metagenomic analysis, and this could favor the growth of some bacterial groups, specifically of the Gammaproteobacteria class. (39)However, this is a controversial discussion since other studies confirmed that the phylogenetic structure and diversity of communities were not significantly influenced by storage temperature or duration of storage. (39,40)Lauber et al. (40) have demonstrated that the relative abundances of most taxa were largely unaffected by temperature even after 14 days of storage in a saline buffer in a study on assessing bacterial community structure in soil and human-skin-associated samples.Moreover, the bacterial relative abundance data in these samples are consistent and comparable to other metagenomic and microbiome studies in indoor environments and transportation facilities with a high circulation of people. (24,36)n the middle of the expedition (stage 2), it was also possible to observe a greater abundance of species in the Eukarya domain, especially fungi of the Ascomycota phyla.The expected higher humidity at this stage of the voyage and different activities on the vessel, including collecting water samples, crew embarking and disem- barking on the sea, food cooking, and use of restrooms, among others, might allow for a more significant proliferation of fungi.Unfortunately, measuring the humidity inside the ship during the collection periods was impossible.Still, the average temperature inside the ship during the voyage ranges from 20 to 25ºC, which can also be an appropriate temperature for the growth of many species of fungi.Genera of yeasts (Scheffersomyces and Debaryomyces) were the most abundant in this stage, followed by Candida.Ascomycetous yeasts are widely distributed in nature, and most are saprotrophs and represent important decomposers, but some species are pathogens of plants and animals as well.Yeasts are generally considered to be mesophiles, and optimal growth temperatures are around 25ºC.Candida represents a fungi genus that colonises the human organism without causing infections.Some are opportunistic pathogens, like Candida albicans, for example.
The abundance of fungi in stages 1 and 3 was shallow concerning stage 2. The presence of fungi at the beginning and end of the expedition could also be related to the temperature range (average temperature in Rio de Janeiro in October 2019 and April 2020 varied between 20 to 25ºC and 25 to 30ºC, respectively) (INMET -Instituto Nacional de Meteorologia -https://clima.inmet.gov.br/progt), thus favouring the proliferation of other fungi species found mainly on surfaces, including molds in stored food or residues and wood, among other surfaces.We collected samples from the air conditioning system, which could also be an essential source of fungi.These microorganisms are usually filtered from the ventilation system to prevent particles and microbes from entering the indoor air. (41,42)However, Cladosporium, Penicillium, and Aspergillus have been detected on passenger ships. (42)nfortunately, a limitation of the study is that we did not perform assays with individual samples collected in the kitchen, bathroom, bedroom, and laboratory areas.Therefore, we could not accurately observe the microbiome characteristics of each surface and compartment.
However, the techniques allowed us to observe some bacterial species related to the Antarctic continent during the different sampling stages.These bacterial species observed are part of the environmental microbiome, not being characterised as potential animal or human pathogens.In addition, the abundance of these species was very low considering the microbiome's total fraction.Studies on the viability of these microorganisms would be necessary to assess whether there is any risk of dispersion to other continents, adaptation, and future ecological impact scenarios.
Virome -Most of the studies performed to evaluate the microbiome of internal surfaces of transportation facilities are directed towards analysing the taxonomic composition of bacteria using 16S rRNA gene sequencing. (6,23)The shotgun metagenome can explore all available DNA in a sample without a specific target.In our study, the methodology used was also directed to capture DNA and RNA, focusing mainly on detecting viruses with RNA genome.To date, few studies on the microbiome characterisation on indoor surfaces in transportation facilities or built environments have focused on the analysis of RNA viruses, and much knowledge about their distribution and patterns of occurrence in the environment is still needed. (6,7,8,23,24)irome analysis demonstrated that RNA viruses prevailed over DNA viruses in the metagenomic libraries.In stage 1 it was possible to identify Pahexavirus that infect bacteria of the genus Propionibacterium sp., which are pleomorphic, occasionally branching bacilli that are the normal flora of the skin, conjunctiva, external ear canal, and exposed mucous membranes. (43)Single-strand ssRNA+ viruses belonging to the Nudiviridae and Polydnaviriformidae families recognised to infect insects were also detected in this stage.One contig in this sample was assigned to NODE 2838 -Actinomyces_virus_Av1 (Podoviridae family) commonly found in humans' mouths.These findings indicate that insect-infecting viruses and commensal microorganisms of the human skin or mucous membranes prevailed on the sampled surfaces without viral pathogens that cause human diseases.The low number of people circulating inside the ship during this sampling period and the previous cleaning of the vessel before receiving the crew and researchers to initiate the expedition could explain the results.
A higher abundance of viral hits and a larger variety of assigned species was observed in pool 2 (swab samples collected in the vessel during the expedition).The result is consistent with many co-livings on the ship (military crews and researchers).In this context, even with the routine cleaning and hygienic measures of the ship´s com- *N50 -length such that sequence contigs of this length or longer include half the bases of the assembly; **L50 -number of sequences contigs that are longer than, or equal to, the N50 length and therefore include half the bases of the assembly.

TABLE V
Viral species (contigs) identified in the pooled samples from different sampling periods partments and surfaces, the large circulation of people could explain a greater spread of viruses in these environments.No known human pathogenic viruses were identified in this stage of the trip.The absence of such viruses was corroborated by the lack of any report of disease outbreaks, such as gastroenteric or respiratory diseases, frequently reported in such confined settings. (2,5,7,9) diversity of bacteriophages within the Siphoviridae family infecting commensal skin bacteria was detected, including Staphylococcus sp., which is genera of Gram-positive bacteria and is part of the normal microbiota of the mouth, skin, intestine, or upper respiratory tract.Bacteriophages, including Staphylococcus and Propionibacterium phages, were also observed in a similar study using shotgun metagenomic analysis to characterise the microbiome of an enclosed public transport (aircraft), where a low abundance of human viral pathogens was also reported. (26)embers of the Virgaviridae family were dominant in pool 2, mainly represented by Tobacco mosaic virus (TMV) species.TMV has an extensive host range, and it is known to infect members of several plant species, including tobacco, tomato, pepper (Solanaceae family), cucumbers, and ornamental flowers, among others.Other viruses infecting plants, vegetables, or other foods, such as Pepper Mild Mottle Virus (PMMoV), Tomato Brown Rugose Fruit Virus, and Watermelon Mosaic Virus, were also detected during this expedition stage.
Among these detected viruses, PMMoV (Virgaviridae family) is abundant in the human gut, consequently in feces and raw sewage and is considered an essential human fecal viral marker in aquatic environments. (44,45,46)owever, we cannot confirm the location of contamination caused by PMMoV due to the pooling of samples.Similarly, this voyage stage also detected a Gemykrogvirus (Contig 526 -Sewage-associated gemycircularvirus 4).The presence of plant-infecting viruses, such as PMMoV, was expected, particularly considering that several sampling sites were close to the crew's dining and kitchen areas.Moreover, due to the high ingestion during regular diet and faecal excretion of PMMoV, studies have considered this virus as an excellent viral marker of human faecal contamination in the environment. (45,46)nitially, to screen for human faecal contamination using viral markers, we screened the 91 individual swab samples collected throughout the expedition for human adenovirus (HAdV) using a qPCR protocol. (47)HAdVs are important viral markers for assessing human faecal environmental contamination. (45)None of the samples tested positive for HAdV.Similarly, by the metagenomic approach, we have yet to identify HAdVs.
In the indoor surface samples, it was possible to identify only one viral species that is probably native to the Antarctic continent (Psychrobacter phage Psymv2) (NCBI RefSeq YP_009017594.1).This bacteriophage was initially isolated from a bacterium of the genus Psychrobacter (Psychrobacter sp.MV2) identified in soil samples in Miers Valley, in the McMurdo Dry Valleys, South Victoria Land, Antarctica. (48)Subsequently, Psychrobacter phage Psymv2 was found to be the most abundant viral species observed in surface and bottom sea samples from Prydz Bay viromes (Antarctica) in 2015. (49)The host bacteria strains of this phage include some members of the genus Psychrobacter, which have been isolated from a wide range of habitats, including surface and deep-sea waters, deep-sea sediments and soil, especially from the Antarctic region, and are also widespread in cold Antarctic environments. (48,49)At the end of the expedition, they were not identified.
In stage 3, occurred the predominance of viruses with a dsRNA genome that infects fungal species of the Penicillium genus.Penicillium was the second genus of fungi most abundant among eukaryotes in this stage of the trip, according to microbiome analysis.
This study documents a first screening to assess the virome at different stages of an expedition to Antarctica, tracking viral communities in a closed environment.Swab-based methods have been used to explore viral contamination on different types of surfaces using RT-qPCR. (50)In contrast, the sensitive next-generation sequencing (NGS) technique and shotgun metagenomic to detect specific pathogenic viruses in surface samples or indoor environments still need to be thoroughly evaluated. (24,26)In a study aimed to investigate the microbiome in urban transit systems, including subways and buses in several cities around the world, researchers have not reported archaea or viruses in such samples. (24)Some limitations for the absence of viral detection in these transport facilities were attributed to the DNA extraction methods used, limitations in sequencing depth, or missing annotations in reference databases used for classification, highlighting the challenges for obtaining the virome in these environments. (24)Moreover, new sampling devices have been tested for microbial sampling, including upgraded swabs (different materials), polyester wipes, macrofoam sponges, adhesive tapes, biological sampling kit (BiSKit; macrofoam), witness coupons, dust, and bulk sampling which could be more effective in concentrating and detecting the microbial population found in indoor surfaces. (51)n our study, frequently cleaning the navy vessel may reduce the presence of pathogenic viruses.Nevertheless, it is essential to emphasise that the absence of viral enrichment protocols can reduce the scope of viral assessment to only the most abundant viruses in the samples.Therefore, viral enrichment protocols can help observe viruses present in lesser abundance in the samples, (52) but it limits the microbiome study by excluding other types of microorganisms.Target-specific detection techniques could be used as a supplementary approach in pathogen surveillance to analyse known pathogens.This approach could be beneficial, especially in the context of pandemic situations.
Our analysis started in October 2019, before the CO-VID-19 pandemic.Therefore, future virome analyses coupled with the search for specific pathogenic viruses could be an exciting strategy to track viral diversity, especially for viruses of public health importance.
Final considerations -Although we did not identify a high abundance of pathogenic microorganisms, it is essential to emphasise that their detection through (meta)genomic analysis does not indicate a direct risk to humans or the environment.Further work is necessary to assess viability and risk of infection.The health risk is associated with several factors, including stability on fomites, the number of infectious agents contacted by the fingerpad, the efficiency of self-inoculation (i.e., transfer of the pathogen from fingerpad to the mouth, nasal cavity, eyes, or other bodily location where infection may occur), the infectious dose of the organism and the individual's susceptibility. (1,16)Therefore, all these variables should be considered in a health risk analysis.
Cleaning and disinfection of contaminated surfaces are frequently implemented measures to control the transmission of pathogens in indoor environments and reduce human fingerprints. (53,54)Still, microorganisms that are sporulated or that form cysts, like some protozoans species, may be more resistant to disinfection.The risk of infection can also be reduced by increasing ventilation in some locations when possible. (11)The use of high-efficiency particulate air filters and ultraviolet germicidal treatment in the ventilation system are significant effective measures. (11,42)In addition, implementing a comprehensive outbreak prevention and control strategy could reduce the impact of viral infection on vessels, particularly relevant in pandemic situations. (13)espite some limitations, this study demonstrates that shotgun metagenomics may be appropriate to describe the microbial diversity of indoor surfaces of transportation facilities, such as cruise or navy ships.Moreover, results suggest that microbial communities on ship's indoor surfaces contain a metapopulation of human skin/mucous commensals and environmental generalists, with variations corresponding to the number of passengers, activities on board, and environmental exposures.This study also evaluated and tracked the abundance of autochthonous or allochthones environmental species between continents and crew, particularly relevant in the Antarctic context.ing; JPGL -funding acquisition, writing -review and editing; MPM -conceptualisation, resources, supervision, funding acquisition, writing -review and editing.All authors have read and approved the final version of the manuscript.No conflict of interest declared.

Fig. 1 :
Fig. 1: swab samples obtained from different surfaces during three stages of a Brazilian Antarctic expedition (Brazilian Navy Polar vessel -Almirante Maximiano H41 -October 2019 to April 2020).

Fig. 2 :
Fig. 2: (A) Relative abundance (%) of bacterial phyla present in each pooled sample of indoor surfaces of the ship during different stages of an Antarctic expedition (2019/2020).(Stage 1: before the expedition; Stage 2: during the expedition; Stage 3: end of the expedition).(B) Relative abundance (%) of bacterial families identified in each pooled sample collected on indoor surfaces of a Navy Polar ship (2019/2020).

Fig. 3 :
Fig. 3: (A) Percentage (%) of fungi phyla (Basidiomycota and Ascomycota) obtained in each metagenomic library concerning total reads assigned.(B) Relative abundance of fungi genera obtained in each expedition stage of the Navy Polar ship (2019/2020).

Fig. 4 :
Fig. 4: heatmap profile showing the relative abundance of viral families detected in each pooled sample of the ship's indoor surfaces during different expedition stages to Antarctica (2019/2020).Each cell contains the number of reads that passed all the selection criteria.Data spanned from white (low relative abundance) to dark blue (high relative abundance), as illustrated by the color scale (log 10 ).

*
(1, 2 and 3) in indoor surfaces from the Brazilian Polar ship during the XXXVIII Brazilian Antarctic expedition, 2019/2020 Contigs identified based on best score and lower e-value through Blastx (search protein database).

TABLE III Taxonomic
classification of bacterial hits related to the Antarctic continent obtained in swab samples collected on indoor surfaces of an Antarctic research ship during different stages of the expedition in 2019/2020 Obs.: Taxonomic annotation of bacterial species with less than 10 reads were not included in the analysis.Stages 1: the beginning of the expedition; 2: middle of the expedition; 3: end of the expedition.*Information retrieved from National Centre for Biotechnology Information (NCBI) accession numbers where the closest related sequences were obtained through Blastx (search protein databases) search.

TABLE IV
De novo assembly: total number of contigs, distribution by length and similarity search using Blastx for viral contigs in each pooled sample from indoor surfaces of a Navy Polar ship during the Antarctic expedition 2019/2020